RAISING HER VOICE Agency and Aspirations of Adolescent Girls on Marriage

EVIDENCE FROM JHARKHAND, INDIA The International Center for Research for Women (ICRW) is a global research institute, with regional hubs in Washington D.C., United States; New Delhi, India; Kampala, Uganda; and Nairobi, Kenya. Established in 1976, ICRW conducts research to identify practical, actionable solutions to advance the economic and social status of women and girls around the world.

ICRW Asia’s thematic focal areas include access to education and livelihoods, adolescent empowerment, gender-based violence, masculinities, gender inequitable attitudes, HIV prevention, and violence against women and girls.

For more information, please visit www.icrw.org/asia

Cover Image: Paula Bronstein/Getty Images/Images of Empowerment

Suggested Citation Achyut, Pranita; Khurana, Nalini V.; Reddy, Hanimi; Gautam, Abhishek & Verma, Ravi K. (2020). Raising Her Voice: Agency and Aspirations of Adolescent Girls on Marriage — Evidence from Jharkhand, India. New Delhi: International Center for Research on Women.

Publication Rights The research reported in this publication has been conducted as part of the UMANG program. UMANG is being implemented by ICRW Asia, in partnership with SATHEE, Badlao Foundation and Project Concern International, in close association with the Government of Jharkhand, and with financial support from the IKEA Foundation. The facts and information in this report may be reproduced/quoted/cited only for non- commercial use and with appropriate attribution. RAISING HER VOICE Agency and Aspirations of Adolescent Girls on Marriage

EVIDENCE FROM JHARKHAND, INDIA

Pranita Achyut, Nalini V. Khurana, Hanimi Reddy, Abhishek Gautam and Ravi K. Verma

2 Contents

Abbreviations 4

Introduction 5

Evaluation Design 7

Findings 9

Discussion 14

References 18

Annexure 19

Acknowledgements 30

3 Abbreviations

CAPI: Computer Assisted Personal Interviewing

ECFM: Early, child and forced marriage

ICRW: International Center for Research on Women

IIPS: International Institute for Population Sciences

OBCs: Other Backward Castes

OR: Odds Ratio

PPS: Probability Proportional to Size

SATHEE: Society for Advancement in Tribes, Health, Education and Environment

SC: Scheduled Caste

SDG: Sustainable Development Goal

ST: Scheduled Tribe

4 Source: A scene from the film, Parvaaz | Flight, produced by ICRW and directed by Mixed Media Productions

INTRODUCTION

Early, child and forced marriage (ECFM) is a violation of a girl’s right to grow in a safe environment, receive education and realize her full potential. It compromises all efforts to overcome poverty, fight for gender equality, advance girls’ education and improve their overall health and well-being.

Several studies strongly suggest that unequal gender norms provide social sanctions and justifications to sustain child marriage practices (Malhotra, Warner, McGonagale, & Lee-Rife, 2011; ICRW, 2008). Child marriage affects a girl’s health, in terms of maternal morbidity, mortality and nutritional status, and has an inter-generational impact on the survival and health of her children (Prakash, Singh, Pathak, & Parasuraman, 2011). ECFM is also linked to a higher probability of girls experiencing intimate partner violence, thereby making it less likely for them to seek and receive the necessary support (Speizer & Pearson, 2011). The status and consequences of ECFM are well recognized, globally and in India.

According to a study conducted by the International Center for Research on Women (ICRW) across multiple countries, a girl’s education is the strongest predictor of the age at which she will marry (Jain & Kurz, 2006).1 Reducing child marriage is a direct target of Sustainable Development Goal 5.3 (SDG 5.3) and it is linked with the achievement of at least eight of the 17 SDGs.2 At the national level, India already has legislation

1Countries included in the analysis are Niger, Chad, Bangladesh, Mali, Guinea, Central African Republic, Nepal, Mozambique, Uganda, Burkina Faso, India, Ethiopia, Liberia, Cameroon, Malawi, Nicaragua, Nigeria and Zambia.

2Goal 1: No poverty, Goal 2: Zero hunger, Goal 3: Good health and wellbeing; Goal 4: Inclusive and quality education, Goal 5: Gender equality, Goal 8: Economic growth, Goal 10: Reduce inequalities, and Goal 16: Peace, justice and strong institutions.

5 prohibiting child marriage as well as multiple programs and schemes to address the issue. Although the prevalence of child marriage has declined, still one out of every four women in the country (aged 20-24 years) were married off before the age of 18 (IIPS and ICF, 2017). This proportion is much higher in the state of Jharkhand (38 percent), particularly in the districts of Godda (63 percent) and Jamtara (44 percent).

To reduce the prevalence of child marriage, ICRW has conceptualized a comprehensive multi-layered girls’ empowerment program called UMANG. The program is being implemented by ICRW in partnership with SATHEE, Badlao Foundation and Project Concern International, and in close association with the Government of Jharkhand. UMANG is being executed with financial support from the IKEA Foundation. The program will reach around 200,000 adolescent girls, men and boys, and community members from four blocks of Godda and Jamtara districts over the four-year program period.

UMANG uses a socio-ecological framework and gender-transformative approaches. It involves multi- layered intervention at individual (adolescent girls), family (parents, brothers/husbands), and community level (men and boys, women, and other community members). The program also works with systems and institutions such as schools, local governance structures, child protection mechanisms, and education and health departments to bring policy and normative shifts. The socio-ecological approach ensures that while adolescent girls are placed at the center of the program, a broader enabling environment for their empowerment is created through engagement across the aforementioned levels.

This research report presents findings from the UMANG program’s baseline survey on marriage practices among adolescent girls aged 15-18 years, and their aspirations and role in decision-making with respect to their marriage. The report also identifies risk and protective factors for ECFM and discusses programmatic implications of the same.

Key Findings

™™ Among girls aged 15-18 years, 11 percent were married and marriage was fixed for another 12 percent. ™™ Out-of-school girls are 3.4 times more likely to be married or have their marriage fixed than girls still attending school. ™™ Among married girls, the average age at marriage was 16 years, approximately two years earlier than when they would have liked to marry. ™™ Among unmarried girls, the desired mean age at marriage is 20 years. However, given their current situation, they thought they will get married a year before their desired age. ™™ Most unmarried girls perceived that they have little say in marriage-related matters and decisions, particularly with respect to the timing (86 percent) and choice of partner (87 percent). ™™ Overall, 82 percent of girls opined that their father is the main decision-maker on matters of marriage. ™™ Most girls (80 percent) recognized the negative implications of early marriage, and supported the need for attaining education and achieving career aspirations before marriage. However, only 43 percent agreed that it is appropriate for girls to express dissent if they do not like their parents’ choice of partner. ™™ When asked to rank education, marriage, physical safety and employment in order of priority for their daughters, parents ranked higher education as a priority over the other three options. At the same time, only 30 percent of parents opined that girls should pursue education up till graduation or above. ™™ About 62 percent of parents felt that if a girl wants to marry a boy of her choice, the family should allow her to marry him. However, 92 percent of parents preferred that their own daughter(s) should marry a boy of their/ family’s choice. ™™ Girls’ school-going status, parents’ education and occupation, caste, religion, and place of residence have varying linkages to girls’ marital status, age at marriage and decision-making related to marriage.

6 Source: Paula Bronstein/Getty Images/ Images of Empowerment

EVALUATION DESIGN

To assess the key outcome indicators related to girls’ empowerment and child marriage, UMANG uses a two-arm (intervention and control) quasi-experimental evaluation design, with baseline and endline data collection. The intervention arm includes four blocks — Mahagama and Godda blocks of Godda district, and Nala and Jamtara blocks of Jamtara district. Program interventions are being implemented in these blocks as part of an integrated approach. Barharwa and Barhait blocks of Sahibganj district have been selected for the control arm, based on their socio-economic and demographic profiles.

The baseline survey was conducted with adolescent girls (aged 10-14 years and 15-18 years), boys and men (aged 12-21 years), and parents (mothers or fathers) of adolescent girls (aged 10-18 years). A sample size of 400 was estimated for each group to measure the outcome indicators separately for each respondent category (girls aged 10-14 years and 15-18 years, boys and men, and parents) at the block level. A two-staged stratified systematic random sampling technique was used for selecting eligible respondents (refer to the Annexure for details on sample size and sampling technique).

7 Overall, more than 8,000 adolescent girls, boys, and parents participated in the baseline survey. Face-to- face interviews were conducted using the Computer Assisted Personal Interviewing (CAPI) technique. Data on marriage, education and employment status, aspirations, attitudes toward gender norms, mobility, communication, and other socio-economic and demographic indicators were collected.

UMANG Program Outcomes

™™ Enhanced aspirations of girls to delay marriage, pursue higher education and seek employment. ™™ Enhanced agency of girls to negotiate the timing of marriage as well as choice of partner, pursue higher education and seek employment. ™™ Positive shifts in attitudes toward gender and marriage-related norms among girls, boys, parents and the community. ™™ Increased inter-departmental convergence on evidence-based planning for adolescents’ empowerment.

ANALYSIS

This report includes data gathered from 2,026 adolescent girls, aged 15-18 years, from six blocks — four intervention and two control— and presents findings on their marital status, age at the time of marriage, aspirations, and involvement in decision-making around marriage. Parents’ perceptions (n=2,049) on ECFM is also included.

For the analysis, the participating adolescent girls have been categorized as: 1) those who are married or have ever been married (married); 2) those whose marriage is fixed, but are not yet married (marriage fixed); and 3) those who are neither married nor has their marriage been fixed (never married). Other indicators included in the analysis are:

™™ Age at marriage— For married girls, the aspirational and actual mean age at marriage have been presented. Unmarried girls (i.e., marriage fixed and never married) were asked two questions: i) at what age they would like to marry; and ii) given their current situation, at what age did they think they will get married. Using these questions, their aspirational mean age at marriage (age at which they would like to marry) and their expected mean age at marriage (age at which they thought they would get married) have been presented.

™™ Decision-making— Girls’ involvement in making decisions around their marriage has been measured in terms of major say, which includes “girls taking the decision on” or “girls having a big say” in deciding the timing of their marriage and choice of partner.

™™ Attitudes toward child marriage— This included statements to capture girls’ perceptions on suitable age at marriage, when (in relation to education and career) to get married, consequences of child marriage and appropriateness of expressing dissent toward parents’ choices around marriage.

™™ Parents’ preferences and perceptions— Parents’ opinions on child marriage have also been included.

Bivariate and multivariate analyses were carried out to examine existing marriage-related practices, risk and protective factors affecting girls’ age at marriage, aspirations of girls and marriage-related decision-making.

8 Source: Impact of Conditional Cash Transfers on Girls’ Education/ICRW

FINDINGS

A) MARITAL STATUS

Overall, 11 percent of girls aged 15-18 Figure 1: Proportion of girls who are married years were married and 12 percent or have their marriage fixed (by age) had their marriage fixed. The proportion of such girls was significantly higher among those aged 14% 17-18 years as compared to 15-16-year- olds (Figure 1). Among 15-year-old 12% girls, two percent were married and nine percent had their marriage fixed. 8% Among 18-year-old girls, the proportion 9% 14% 24% increases to 24 percent and 14 percent, 7% respectively. 2% 15 16 17 18 Marital status of girls varied significantly by place of residence. Ever Married Marriage Proposed The proportion of married girls ranged from eight percent in Nala block to

9 16 percent in Godda block (refer to Table 2 in the Annexure). On the other hand, 14 percent of girls in Jamtara block, and six percent in Godda block had their marriages fixed. After adjusting for background characteristics, in comparison to Nala block, girls from other blocks are 1.7 to 1.9 times more likely to be married or have their marriage fixed by the age of 18 (refer to Table 3 in the Annexure).

School-going status of girls strongly influences their marital status and acts as a protective factor from child marriage. Among girls who were in school, 88 percent were neither married nor had their marriage fixed as compared to 68 percent among those who were out of school (refer to Table 2 in the Annexure). From the multivariate analysis, it is evident that out-of-school girls are 3.4 times more likely to be married or have their marriage fixed than those who are still in school (refer to Table 3 in the Annexure).

School dropout preceded marriage for 59 percent of girls. Among married girls, only 16 percent were currently in school as compared to 38 percent of those whose marriage was fixed and 57 percent of those who were neither married nor was their marriage fixed (refer to Table 4 in the Annexure). Among girls who dropped out of school, only nine percent first got married and then dropped out, while 32 percent got married and dropped out in the same year. About 59 percent dropped out at least a year before getting married (Figure 2). The girls cited several reasons for dropping out of school, including education being expensive, schools being inaccessible or too far, and household responsibilities. Once the girls are out of school, parents perceive marriage as a “safe” and preferred option for them.

Figure 2: Timing of girls' marriage (in comparison to timing of school drop-out), among married, school dropped-out girls aged 15-18 years

32% 33%

19%

6% 7% 3%

2 to 5 1 Same 1 to 2 3 to 5 >5 years year year years years years before before after after after

Fathers’ education emerged as a protective factor against marriage of girls. As per the bivariate analysis, a significantly higher proportion of girls with their mother or father having 10 or more years of schooling were neither married nor was their marriage fixed, as compared to girls having illiterate parents or parents with fewer years of schooling (refer to Table 2 in the Annexure). After controlling for other factors, girls with fathers with no schooling (Odds Ratio or OR=1.65) or less than 10 years of schooling (OR=1.56) are more likely to be married or have their marriage fixed than girls whose fathers have 10 or more years of schooling (refer to Table 3 in the Annexure). In the multivariate analysis, mother’s education did not show significant effect.

Influence of parents’ occupation on the marriage of girls was complex. The bivariate analysis revealed that a higher proportion of girls whose mothers are engaged in agricultural or unskilled labor were married, or their marriage was fixed (28 percent) as compared to girls whose mothers are not working (19 percent). The proportion of married girls or those whose marriage was fixed was the lowest among girls whose fathers are cultivators or non-agricultural skilled workers (18 percent) than those who are unskilled workers (23 percent) or engaged in business or salaried jobs (25 percent) (refer to Table 2 in the Annexure). When

10 adjusted for other background characteristics in the multivariate analysis, the occupation of mothers and fathers showed different effects on the marital status of girls. Girls with mothers who are not working are less likely to be married or have their marriage fixed by the age of 18 years (OR=0.68), as compared to those who are unskilled workers. On the other hand, girls with fathers engaged in business or salaried jobs are 1.5 times more likely to be married or have their marriage fixed as compared to girls whose fathers are engaged in unskilled work or not working at all (refer to Table 3 in the Annexure).

Religion had an influence on the timing of marriage. When adjusted for other factors in the multivariate analysis, it was noted that girls from Hindu families are 1.6 times more likely to be married than those who were from non-Hindu families (refer to Table 3 in the Annexure).

B) AGE AT MARRIAGE: DESIRED AND ACTUAL

Figure 3: Age at Marriage: Actual, likely and desired

25

20.5 20 18.2 19.3

15.9 15

10

5

0

Actual age Desired age Likely age Desired age at marriage at marriage at marriage at marriage

Married Girls Unmarried Girls

Among married girls, the mean age at marriage was 16 years. This is two years lower than the age at which the girls wanted to be married. The mean age at marriage was the lowest in Godda block (15.4 years), while the highest was in Nala block (16.3 years) (refer to Table 5 in the Annexure). Girls in Nala block also reported a higher desired age at marriage (18.8 years).

Unmarried girls aspired to get married at least a year later than when they thought they were likely to get married. The desired mean age at marriage was 20.5 years, while the likely mean age at marriage (as perceived by girls) was 19.3 years (refer to Table 5 in the Annexure). Nearly half of the girls (45 percent) mentioned that they thought they were likely to get married at the age they wish to marry, while 29 percent thought that they were likely to be married a year or two before their desired age. Another 20 percent thought that they would be married three or more years earlier than their desired age.

Among unmarried girls, those who were in school and had literate parents were more likely to aspire to be married at a later age. School-going girls were two times more likely to express their desired age at marriage to be more than 20 years than those out of school, after adjusting for other background characteristics (refer to Table 6 in the Annexure). Further, older girls (17-18 years) were more likely to have a higher desired age at marriage than younger ones (15-16 years). Girls whose mothers have attended school

11 are more likely (OR=1.5) to aspire for a higher age at marriage than those with mothers who have never been to school. Further, girls with fathers with 10 or more years of schooling are 1.8 times more likely to aspire for a higher age at marriage than those who have no schooling. Caste also emerged as a predictor; girls belonging to general castes (castes other than Scheduled Caste (SC), Scheduled Tribe (ST) or Other Backward Castes (OBCs)) are more likely to aspire for a higher age at marriage than those from STs.

C) GIRLS’ INVOLVEMENT IN DECISION-MAKING RELATED TO MARRIAGE

Most married girls had limited or no say in the decisions related to their marriage. Only 15 percent of girls reported having had a major say in deciding when and whom to marry, with significant inter-block variation (refer to Table 7 in the Annexure). In Mahagama block, only nine percent of girls reported that they had a major say in deciding the timing of their marriage. The proportion of such girls was 31 percent in Nala block. Similarly, nine percent of girls in Godda block and 23 percent of girls in Nala block reported having a major say in choosing their partner.

Only 13-14 percent of never married girls expressed that they would have a major say in deciding the timing of their marriage and choice of partner. A significantly higher proportion of girls from Nala block (23 percent) believed that they would have a major say in deciding the timing of their marriage as compared to 5 percent of girls in Godda block (refer to Table 8 in the Annexure). Similarly, 20 percent of girls in Nala, 14 percent in Jamtara, 13 percent in Mahgama and five percent in Godda blocks believed that they would have a major say in choosing the partner. Overall, only 22 percent of girls — 35 percent in Jamtara and 14 percent in Godda districts — expressed their desire to meet a potential partner before marriage. Four out of ten girls expressed being scared, anxious or unhappy about marriage, while one third said that they never thought about it.

Fathers had maximum say in arranging their daughters’ marriages. About 81 percent of ever married girls mentioned that their father was the main decision maker (refer to Table 9 in the Annexure). A similar proportion of girls, who were never married, expressed the same.

Decision-making on the timing of marriage varied by block, girls’ school-going status, and fathers’ occupation and religion. Girls from Godda and Mahagama blocks were less likely to perceive that they have a major say in deciding the timing of their marriage as compared to girls from Nala block (refer to Table 10 in the Annexure). School-going girls were more likely to perceive that they would have a major say in deciding the timing of their marriage than those who dropped out, after adjusting for other background characteristics. Notably, a similar association did not emerge for decisions on the choice of partner (refer to Table 11 in the Annexure).

Fathers’ occupation and religion also emerged as predictors of space for girls to make decisions on the timing of their marriage. Girls with fathers who are cultivators or skilled workers are two times more likely to perceive that they have a major say in deciding the timing of their marriage than girls whose fathers are unemployed or unskilled workers. Further, girls from Hindu families are more likely to perceive that they have a major say as compared to girls from other religious groups (OR=1.6).

D) ATTITUDES TOWARD CHILD MARRIAGE

Most unmarried girls mentioned 18 years or more as the right age for girls to get married. They also considered child marriage to be a harmful practice, but did not support dissent. Further, awareness of the law prohibiting child marriage was low. When asked about the right age for marriage of girls, 58 percent of girls mentioned 18 years, while 38 percent mentioned between 19 and 25 years. Around 80 percent of girls agreed or strongly agreed that marrying early could have a negative impact on girls’ education, and that a girl should be married only after she has achieved her educational and career aspirations (refer to Table 12 in the Annexure). Three-fourths of unmarried girls disagreed or strongly

12 Source: Planning Ahead for Girls’ Empowerment and Employability/ICRW.

disagreed with the prevailing justifications for ECFM, namely, marrying girls young could help resolve financial problems in the family, protect them from facing sexual violence and would require a lower dowry than for older brides.

However, only 43 percent of girls agreed or strongly agreed that “it is appropriate for girls to express displeasure if they do not like the partners chosen for them”. Only 43 percent of girls disagreed with the statement that “even if a girl does not want to be married, she should honor the decisions/wishes of her family”.

E) ATTITUDES OF PARENTS TOWARD MARRIAGE

Contrary to prevailing beliefs, the top priority for parents was higher education3 of their daughters followed by their marriage. However, only 30 percent of parents opined that a girl should complete her education up to graduation or above. Parents ranked the physical safety of their daughters third, and their employment/career fourth.

When presented with the following situation — a girl wants to marry a boy of her choice — and asked what the family should do, 62 percent of parents opined that the family should allow them to marry (refer to Table 13 in the Annexure). A third were against letting them marry and five percent expressed that the girl should be quickly married off elsewhere. In Mahagama block, 41 percent were against allowing the girl to marry a boy of her choice, while nine percent suggested that parents should get their daughter married elsewhere soon. On the other hand, in Jamtara block, 71 percent of parents expressed that the girl should be allowed to marry a boy of her choice. However, 92 percent of parents expressed that they would prefer to have their own daughter marry a person of the family or parents’ choice. So, even when parents’ intent is to prioritize education over marriage, they expect their daughters to follow their decisions on when and whom to marry. Further, the right age for marriage for girls as perceived by parents, was lower than what the girls mentioned (18.7 years viz. 19.3 years).

3Parents are likely to consider education beyond the 10th standard as higher education.

13 Source: Paula Bronstein/Getty Images/ Images of Empowerment

DISCUSSION

With concerted efforts, child marriage in India has declined substantially over the last two decades. Yet, it continues to be widely practiced in certain districts and sub-districts. Godda and Jamtara are two such districts in Jharkhand with a high prevalence of child marriage. Further, the pace of decline in child marriage is higher among younger adolescents (10-14 years) than older adolescents (15-18 years) (Srinivasan, et al., 2015). Findings from the UMANG baseline survey are consistent with this, and thus, this report focuses on marriage practices among older adolescent girls.

This baseline study noted that 11 percent of girls aged 15-18 years were already married, while another 12 percent had their marriage fixed. Within this age group, fewer 15-year-old girls were either married or had their marriage fixed as compared to 18-year-old girls. The average age at marriage among 15-18-year- old girls was only 16 years, which is two years below the legal minimum age for marriage of girls in India.

The findings highlight that girls in the selected districts for this study are likely to be married early, i.e., before their desired age for marriage. On an average, married girls had been married off approximately two years before they wished to marry, while unmarried girls were likely to be married off one year before they wish to marry. Further, the findings reveal that unmarried girls’ desired age for marriage was two years higher than

14 that of married girls, suggesting that unmarried girls aspired for a delayed marriage. Conversely, those girls who were married at a young age may express a lower desired age for marriage due to a constrained sense of aspirations within their marriage.

The gap between the girls’ actual or expected age at marriage and their desired age at marriage raises the question: why are girls getting married or are likely to get married before they want to? Existing research indicates that this may be because girls have little or no space to take decisions related to their marriage, with married girls experiencing even greater restrictions on exercising their agency (Dasra, 2019; Breakthrough, 2013). These findings are echoed in this baseline study. Only five percent of married girls and 1.5 percent of unmarried girls shared that they themselves had or would have a significant say in marriage- related decisions.

This indicates that for most girls aged 15-18 years, marriage is not a matter of choice but a mandatory event in their lives, which they feel is beyond their control. This implies a lack of agency and choice. This also reflects on their feelings toward marriage. Four out of 10 unmarried girls expressed feeling scared, anxious, or unhappy about marriage, while one-third expressed that they had never thought about it. This could be linked with adverse mental health and wellbeing outcomes, as found in ICRW’s study in Niger and Ethiopia (ICRW, 2018; ICRW, 2019), indicating the need for further study and exploration in this context.

Fathers of adolescent girls emerged as key decision-makers with respect to their marriage. While parents, in principle, agreed that a girl should be allowed to marry a partner of her choice, they were unwilling to apply this to their own daughters. Various studies have linked the central role of fathers in decision-making regarding their daughters’ marriage with prevailing norms around masculinity, honor, and sexuality (Nirantar Trust, 2015; Breakthrough, 2013). A girl’s chastity and protection are considered important markers of the father’s honor and masculinity, and losing control over his daughter can lead to his ostracization from the community. This is because marriage is seen as a tie between two families — an agreement so significant that only fathers/male family members are entrusted with the decision, while mothers lend their support without questioning the decision nor making an independent assessment of the partner chosen for their daughter.

Parents’ inflexibility regarding the choice of partner is closely related to norms around “caste purity” and marriage within caste or religious groups; thereby enabling their control over women and girls’ sexuality. As a result, parents — particularly fathers — strictly control decisions pertaining to the marriage of their daughters to ensure compliance with prevalent norms and eliminate the risk of transgression.

Girls are not only excluded from the decision-making process related to their own marriage, but also do not have the space to voice any dissent against their father’s or parents’ decisions. As per the survey, though most girls opposed early marriage and recognized its harmful impacts, more than half of them did not consider it appropriate to express dissent. They believed that girls must adhere to the family’s decision, even if it is not aligned with their desires, aspirations and choice. Deep-rooted norms related to respect and obeyance of parents, expectations of being a “good daughter”, and notions of honor linked to girls’ marriage, together create a situation in which girls lack not only choice, but also an ability to voice their opinions freely and make (or at least have a major say in) key decisions related to their marriage.

Existing research has demonstrated a positive link between the education of girls and a delayed marriage (ICRW, 2008; Greene & Stiefvate, 2019). It has also highlighted the complex, bi-directional relationship between child marriage and education. Given prevailing gender norms and the low value of girls, marriage and education are often seen as competing priorities in the lives of adolescent girls. Therefore, girls may drop out of school due to their marriage being fixed, or girls who have dropped out may become more

15 Source: Planning Ahead for Girls’ Empowerment and Employability/ICRW.

vulnerable to marriage in the proximate future. Findings from the UMANG baseline study support the latter, i.e., girls who are in school are significantly less likely to be married or have their marriage fixed. Once out of school, girls face a heightened vulnerability to marriage as they begin to receive proposals from prospective partners’ families and matchmakers. Given the lack of alternatives, parents find marriage to be a safe option for girls and get them married off (Breakthrough, 2013).

School-going girls are also more likely to aspire for marriage at a later age, and perceive having greater choice in decisions related to the timing of their marriage. However, no significant link could be found between a girl’s schooling status and the perceived say in choosing the partner. Irrespective of schooling status, girls lack voice and choice in partner selection, and this decision rests entirely in the hands of their fathers/parents.

This baseline study revealed the inter-generational impact of parents’ education on the lives of their daughters and the extent of their aspirations. Girls with parents who have completed a lower level of education are more likely to be married or have their marriage fixed, and have a lower desired age for marriage as compared to girls with parents educated up to a higher level. However, the occupation of parents has a complex relationship with the marital status of their daughters. Girls with fathers engaged in occupations of higher status and higher pay were more likely to be married. On the other hand, girls with mothers who were not working (which can be linked to higher family wealth and privilege as compared to mothers engaged in unskilled or agricultural labor) were less likely to be married.

Despite these contradictory implications, the study found no significant or direct link between family wealth and the likelihood of a girl being married or having her marriage fixed, in the given population. Existing research and learnings suggest that the relationship between affluence and child marriage, too, is bi- directional (Breakthrough, 2013). On the one hand, circumstances such as poverty and inability to pay for education may be an enabler for child marriage, as a way to reduce the burden of providing for daughters. On the other hand, wealthier families are able to find partners for their daughters easily and can pay higher dowries, resulting in the urgency to marry off the daughter to the most affluent possible match.

16 These findings have important implications for policy and programming on ECFM. Recognizing the protective effect of schooling, the UMANG program includes school-based interventions with girls and boys in standards 6-12. Participatory group education activities are held with adolescents that build perspective on gender, power, and other key concepts, and enhance adolescents’ skills of communication, negotiation, and decision-making. These life skills are particularly important to enhance girls’ agency and enable them to bridge the gap between their current situation and their aspirations related to education and marriage. The school-level intervention also includes student-led campaigns and engagement with institutional platforms such as School Management Committees and Bal Sansad (children’s/youth parliament) to create an enabling environment for larger change and sustainability.

Given the heightened vulnerability of girls who have dropped out from school to ECFM, UMANG also includes community-based interventions with adolescent girls. These interventions use sports as a medium to enhance girls’ agency and aspirations, supplemented by group education activities to build their perspective and life skills, with the aim to bring them back to school and delay their marriage.

The findings also highlight the central role played by men, particularly fathers, in making key decisions related to their daughters’ education and marriage. The UMANG program thus includes a specific component to engage men and boys, which is youth-led and uses formal and informal structures and platforms including local governance, sports clubs, cultural groups, and other congregation points to initiate dialogue and call for action. The aim is to sensitize men and boys toward gender discrimination and create an enabling environment, wherein girls can exercise their voice and choice in key matters pertaining to their life and wellbeing.

With growing awareness and aspirations among girls and parents to prioritize education over marriage, there are opportunities for engaging different stakeholders to challenge and change ECFM-related practices. UMANG recognizes that girls, parents, and community members are embedded in a larger normative system and a part of key structures and institutions. Thus, the program also engages with institutions and structures, and positions different intervention components within the system for larger and sustained change.

Source: A scene from the film, Parvaaz | Flight, produced by ICRW and directed by Mixed Media Productions

17 REFERENCES

Breakthrough. (2013). Marriage Can Wait. Our Rights Can’t. The Causes, Consequences, and Resistance of Early Marriage in Bihar and Jharkhand. Breakthrough.

Das Gupta, S., Mukherjee, S., Singh, S., Pande, R., & Basu, S. (2011). Knot Ready: Lessons from India on Delaying Marriage for Girls. ICRW.

Dasra. (2019). Situation of Adolescents in Jharkhand: Topline findings from a state-wide survey. New Delhi: Dasra. Retrieved from https://www.dasra.org/assets/uploads/resources/Situation%20of%20Adolescents%20in%20 Jharkhand-%20Topline%20Findings%202019-HiRes.pdf

Greene, M. E., & Stiefvate, E. (2019). Social and gender norms and child marriage: A reflection on issues, evidence and areas of inquiry in the field. London: ALIGN. Retrieved from https://www.alignplatform.org/resources/2019/04/ socialgender-norms-and-child-marriage

ICRW. (2008). Knot Ready: Lessons from India on Delaying Marriage for Girls. ICRW.

ICRW. (2018). A Life Not Chosen: Early Marriage and Mental Health. ICRW. Retrieved from https://www.icrw.org/ publications/a-life-not-chosen-early-marriage-and-mental-health/

ICRW. (2019). Child Marriage, Intimate Partner Violence & Mental Health Among Young Ethiopian Women. ICRW.

International Institute for Population Sciences (IIPS) and ICF. (2017). National Family Health Survey (NFHS-4), 2015-16. Mumbai: IIPS.

Jain, S., & Kurz, K. (2006). ICRW research on prevalence and predictors of child marriage in developing countries. Retrieved from https://www.icrw.org/wp-content/uploads/2016/10/Child-Marriage-Toolkit.pdf

Malhotra, A., Warner, A., McGonagale, A., & Lee-Rife, S. (2011). Solutions to End Child Marriage: What the Evidence Shows. ICRW.

Nirantar Trust. (2015). Early and Child Marriage in India: A Landscape Analysis. Nirantar Trust.

Prakash, R., Singh, A., Pathak, P. K., & Parasuraman, S. (2011, July). Early marriage, poor reproductive health status of mother and child well-being in India. Journal of Family Planning and Reproductive Care , 37 (3), pp. 136-45.

Speizer, I. S., & Pearson, E. (2011). Association between Early Marriage and Intimate Partner Violence in India: A Focus on Youth from Bihar and Rajasthan. J Interpers Violence, 26(10), pp. 1963-81.

Srinivasan, P., Khan, N., Verma, R., Giusti, D., Theis, J., & Chakraborty, S. (2015). District-level study on child marriage in India: What do we know about the prevalence, trends and patterns? New Delhi: ICRW.

18 ANNEXURE

SAMPLING SIZE AND SAMPLING TECHNIQUE:

Sample size: To measure the outcome indicators separately for girls aged 10-14 years and 15-19 years at the block level, the sample size has been estimated using the formula given below:

2 2 n = D*{zα √2*P*(1-P) + Zβ √P1(1-P1) + P2 (1-P2)} /(P1-P2)

Where,

P1 is the baseline value of a specified indicator.

P2 is the end line value of specified indicator (±15% from baseline).

P= Assumed value of the indicator: (P1+P2)/2 n = Sample size

Zα = 95% confidence interval of P value (1.96)

Zβ = 90% power {probability of rejecting hypothesis when it is false (1.28)} D = Design-effect (1.5)

Assuming the base figure to be 50 percent, a sample of 400 is sufficient to detect a 15 percent change from baseline to end line at 95 percent level of significance and 90 percent power, 1.5 design effect and 10 percent non-response.

Sampling technique: A two-staged stratified systematic random sampling technique has been used for selection of eligible respondents.

First stage: Census villages were considered as a Primary Sampling Unit (PSU). Any village with a population of less than 500 persons was linked with a neighboring village to construct a PSU of around 1000 persons. Subsequently, all the villages/PSUs in each block were stratified into the following three tiers:

™™ Tier-1 included villages/PSUs with a government secondary/upper primary or higher school that offers education from standards 6–12.

™™ Tier-2 included villages that are at a moderate distance of 1–5 km from Tier-1 villages.

™™ Tier-3 included all remaining villages, most of which are remote.

Despite a uniform sample size of 400 girls per block, the number of PSUs covered per block were varied to ensure a coverage of at least 15 percent of the villages per block and adjust for the varying mean number of girls available per village per block. The number of PSUs to be selected from each stratum was allocated according to the proportion of population of each of the three strata. Within each strata of a block, the allocated number of PSUs/villages were selected by using a systematic Probability Proportional to Size (PPS) sampling design. The bigger villages (with population more than 1500) were stratified into three or more segments of 500 persons, and from each segmented village, two segments were randomly selected. All the households in the selected PSUs were listed and girls aged 10-14 years and 15-18 years were identified.

Second stage: The list of identified girls constituted the sampling frame for the selection of eligible respondents. The required samples were selected using Systematic Random Sampling.

The baseline survey was conducted during July-August 2019.

19 Table 1: Proposed and achieved sample size per block for 15-18-year-old girls and parents

Area Intervention Control

Godda district Jamtara district Sahibganj district District and blocks Godda Mahagama Jamtara Nala Barhait / Barharwa

15-18-year-old girls

Proposed 400 400 400 400 400

Achieved 402 408 403 406 406

Parents of 10-18-year-old girls

Proposed 400 400 400 400 400

Achieved 402 422 410 411 404

Table 2: Marital status of girls aged 15-18 years by background characteristics

Background Characteristics Ever Married Marriage Fixed Never Married Weighted Count

Block** Mahagama 10.5% 11.7% 77.7% 408

Godda 16.3% 6.4% 77.3% 402

Jamtara 11.5% 13.5% 74.9% 403

Nala 8.4% 8.1% 83.5% 406

Barhait 14.4% 11.7% 73.9% 107

Barharwa 9.2% 13.2% 77.6% 300

Currently in school*** No 19.3% 13.2% 67.5% 1007

Yes 3.6% 7.9% 88.5% 1019

Age*** 15-16 years 4.6% 8.6% 86.8% 1096

17-18 years 19.5% 12.8% 67.6% 929

Mother’s education** Illiterate 13.0% 10.8% 76.2% 1432

1-9 years 7.6% 10.5% 81.9% 501

10+ years 8.0% 6.6% 85.5% 93

Father’s education*** Illiterate 15.4% 10.4% 74.3% 964

1-9 years 9.2% 11.7% 79.1% 692

10+ years 5.4% 8.7% 86.0% 370

Agricultural labor/Unskilled Mother’s occupation** 13.5% 14.9% 71.6% 423 labor Cultivator/Non-agricultural 11.8% 9.2% 79.0% 735 skilled worker

Not working 10.4% 8.6% 81.0% 637

Business/Salaried 9.6% 11.7% 78.7% 230

Agricultural labor/Unskilled Father’s occupation* 12.2% 10.8% 77.0% 694 labor Cultivator/Non-agricultural 10.3% 8.1% 81.6% 710 skilled worker

20 Background Characteristics Ever Married Marriage Fixed Never Married Weighted Count

Business/Salaried 11.9% 13.0% 75.1% 623

Religion Non-Hindu 10.5% 11.6% 78.0% 789

Hindu 12.1% 9.8% 78.1% 1237

Caste ST 13.9% 8.7% 77.4% 393

SC 14.4% 10.4% 75.1% 250

OBC 10.8% 10.6% 78.7% 1068

Other 8.3% 12.7% 79.0% 315

Type of village School village (T1) 13.0% 11.5% 75.5% 601

Medium distance from T1 10.3% 9.1% 80.6% 660 (T2)

Far off village from T1 (T3) 11.3% 10.9% 77.9% 764

Total (%) -- 11.5 11.9 76.6 2026

Note: Chi Square test, *p<0.05., **p<0.01, ***p<0.001

Table 3: Multivariate logistic regression odds ratio (OR) for girls aged 15-18 years on marital status

Dependent variable – Married or marriage fixed=1; rest=0 95% C.I.

Background characteristics Odds Ratio Lower Upper Sig.

Block - Nala ® .023

Mahagama 1.715 1.105 2.663 .016

Godda 1.940 1.258 2.992 .003

Barhait/Barharwa 1.866 1.197 2.911 .006

Jamtara 1.762 1.152 2.697 .009

School status - Currently in school ®

Not in school 3.442 2.590 4.575 .000

Age – 15-16 years ® 17-18 years 2.751 2.139 3.538 .000

Mother's education - At least 1 year of schooling ® No schooling 1.033 .748 1.427 .843

Father's education - 10 or more years of schooling ® .055 No schooling 1.651 1.075 2.535 .022 1-9 years of schooling 1.565 1.050 2.333 .028

Mother's occupation – Unskilled worker ® Skilled worker/ business/ salaried .825 .596 1.142 .247 Not working .679 .465 .991 .045

Father's occupation – Not working or unskilled worker ® Cultivator/Skilled worker .913 .675 1.233 .553

21 Dependent variable – Married or marriage fixed=1; rest=0 95% C.I.

Background characteristics Odds Ratio Lower Upper Sig.

Business/Salaried 1.522 1.079 2.146 .017

Religion – Non-Hindu ® Hindu 1.573 1.166 2.123 .003

Caste – ST ® SC .767 .471 1.250 .287 OBC .947 .648 1.382 .777 Other .914 .561 1.491 .720

Type of village – T1 ® T2 .703 .512 .965 .029 T3 .891 .657 1.210 .460

Wealth quintile – Richest ® Poorest .743 .480 1.152 .184 Poor .867 .573 1.313 .502 Moderate .829 .556 1.238 .360 Rich .907 .622 1.321 .610

Constant .035 .000 Note: ® — Reference Category

Table 4: Link between school-going status and marital status of girls aged 15-18 years

Marital/School-going status Never been to school Discontinued Currently in school Total (N)

Ever married 7.3 76.7 15.9 232

Marriage fixed 2.3 60.1 37.6 213

Never married 2.4 40.5 57.1 1581

Total (%) 3.0 46.7 50.3 2026

Table 5: Desired and actual or likely age at marriage among married and unmarried 15-18 year old girls

Ever married Unmarried

Desired age at Actual age at marriage Desired age at Likely age at marriage Block marriage (Mean) (Mean) marriage (Mean) (Mean)

Mahagama 18.1 16.1 20.6 19.1

Godda 18.2 15.4 20.2 19.2

Jamtara 18.0 16.1 20.6 19.6

Nala 18.8 16.3 20.7 19.6

Barhait/Barharwa 18.0 16.1 20.2 18.8

Total 18.2 15.9 20.5 19.3

Total (N) 232 232 1718 1380

22 Table 6: Multivariate logistic regression for desired age at marriage to be over 20 years among unmarried girls aged 15-18 years

Dependent variable – desired age at marriage >20 years =1; rest=0 95% C.I. for EXP(B) Exp(B) Sig. Background characteristics Lower Upper

Block - Nala® .494

Mahagama .948 .653 1.377 .781

Godda .807 .558 1.166 .254

Barhait/Barharwa .932 .636 1.366 .718

Jamtara 1.148 .800 1.648 .454

School status – Not in school ®

Currently in school 1.906 1.490 2.437 .000

Age – 15-16 years ®

17-18 years 1.946 1.556 2.432 .000

Mother’s education – No schooling®

At least 1 year of schooling 1.484 1.137 1.935 .004

Father’s education – No schooling ® .000

1-9 years of schooling .897 .685 1.175 .429

10 or more years of schooling 1.829 1.300 2.573 .001

Mother’s occupation – Unskilled worker ®

Skilled worker/business/salaried 1.046 .765 1.430 .780

Not working 1.205 .849 1.710 .297

Father’s occupation – Not working or unskilled worker ®

Cultivator/Skilled worker 1.231 .944 1.606 .125

Business/Salaried 1.139 .833 1.559 .414

Religion – Non-Hindu ®

Hindu 1.283 .981 1.678 .069

Caste – ST ®

SC .761 .480 1.207 .246

OBC 1.024 .729 1.440 .890

Other 1.781 1.162 2.731 .008

Type of village – T1 ®

T2 .852 .645 1.125 .258

T3 .835 .634 1.099 .198

Wealth quintile – Richest ®

Poorest .986 .665 1.460 .942

Poor .943 .653 1.363 .755

Moderate 1.007 .712 1.423 .971

Rich .883 .638 1.223 .455

Constant 1.082 .817

Note: ® — Reference Category

23 Table 7: Extent to which ever married girls aged 15-18 years had a say in decisions related to their marriage

Mahagama Godda Jamtara Nala Barhait/ Barharwa Grand total

Extent to which girls had a say in deciding timing of their marriage (%)

No say 65.1 73.8 45.7 48.6 72.1 62.5

Some say 25.6 16.9 34.8 17.1 20.9 22.8

Big say 4.7 7.7 10.9 20.0 2.3 8.6

Girls took decision 4.7 1.5 10.9 11.4 7.0 6.5

Extent to which girls had a say in choosing their partner (%)

No say 60.5 84.6 60.9 62.9 69.8 69.4

Some say 27.9 7.7 21.7 11.4 16.3 16.4

Big say 7.0 9.2 6.5 11.4 4.7 7.8

Girl took decision 4.7 0.0 13.0 11.4 9.3 6.9

Total (N) 43 65 46 35 43 232

Table 8: Extent to which 15–18 year-old girls who are never married (neither married nor marriage fixed) had a say in decisions related to their marriage and their feelings on marriage

Barhait/ Grand Mahagama Godda Jamtara Nala Barharwa total

Extent to which a girls are likely to have a say in deciding timing of their marriage (%)

No say 62.1 66.9 44.7 41.9 69.6 56.9

Some say 24.3 14.1 31.8 30.1 16.7 23.5

Big say 8.2 3.2 14.9 18.9 9.3 11.0

Girls will take the decision 3.8 1.9 1.7 4.1 2.9 2.9

Have not thought/do not know 1.6 13.8 7.0 5.0 1.6 5.8

Extent to which a girls are likely to have a say in choosing their partner (%)

No say 60.3 72.0 48.3 41.0 67.9 57.7

Some say 24.9 10.0 30.8 31.9 15.4 22.7

Big say 10.4 3.5 12.3 17.1 8.3 10.4

Girl will take the decision 2.8 1.3 1.3 2.9 3.5 2.4

Have not thought/do not know 1.3 13.2 7.3 7.4 4.5 6.7

Feelings about getting married (%)

Excited/Looking forward 12.0 14.1 9.6 15.6 14.0 13.6

Nothing special 15.8 10.3 12.3 12.7 12.9 13.2

Scared 30.6 13.5 13.9 17.4 17.6 19.2

Anxious 11.4 10.3 16.6 17.1 12.1 13.9

Unhappy 2.8 6.8 5.3 9.4 3.0 5.6

Never thought of it 27.1 45.0 42.4 28.0 26.1 34.4

Girls who prefer to meet prospective partner before 15.5 12.2 34.8 34.2 14.7 22.4 marriage (%)

Total (N) 317 311 302 339 312 1581

24 Table 9: Person having maximum say in arranging marriage of girls among those who are ever married and never married (neither married nor marriage fixed)

Mahagama Godda Jamtara Nala Barhait/ Barharwa Grand total

Ever married girls (%)

Girl/respondent 4.7 1.5 10.9 11.4 0.0 5.2

Father 88.4 70.8 76.1 77.1 95.3 80.6

Mother 4.7 18.5 6.5 8.6 4.7 9.5

Other relatives/others 2.3 9.3 6.5 2.9 0.0 4.7

Total number of ever married girls (N) 43 65 46 35 43 232

Never married (%)

Girl/respondent 0.9 0.3 2.6 2.9 0.3 1.5

Father 89.6 77.2 76.2 80.2 87.8 82.2

Mother 7.6 12.5 14.6 10.6 10.6 11.1

Other relatives/others 1.9 7.7 5.3 3.5 1.0 3.9

Total number of never married girls (N) 317 311 302 339 312 1581

Table 10: Multivariate logistic regression on decision-making power on when to marry among unmarried girls

Dependent variable – Girl having a major say on 95% C.I. when to marry=1; rest=0 Sig.

Background characteristics Odds Ratio Lower Upper Block – Nala ®

Mahagama .504 .308 .825 .006

Godda .160 .087 .296 .000

Barhait/Barharwa .827 .511 1.338 .439

Jamtara .880 .564 1.375 .575

School status – Not in school ®

Currently in school 1.589 1.121 2.252 .009

Age – 15–16 years ®

17-18 years 1.161 .855 1.576 .338

Mother's education –No schooling ®

At least 1 year of schooling 1.265 .886 1.807 .196

Father's education – No schooling ®

1-9 years of schooling 1.111 .753 1.641 .596

10 or more years of schooling 1.376 .851 2.224 .193

Mother's occupation – Unskilled worker ®

Skilled worker/business/salaried 1.417 0.853 2.355 .178

Not working 1.563 .906 2.693 .108

Father's occupation – Not working or unskilled worker ®

Cultivator/Skilled worker 2.440 1.633 3.647 .000

Business/Salaried 1.489 .937 2.366 .092

25 Dependent variable – Girl having a major say on 95% C.I. when to marry=1; rest=0 Sig.

Background characteristics Odds Ratio Lower Upper

Religion – Non-Hindu ®

Hindu 1.682 1.147 2.467 0.008

Caste – ST ®

SC 1.527 .844 2.764 .162

OBC .970 .616 1.525 .893

Other 1.127 .622 2.041 0.694

Type of village – T1 ®

T2 1.066 .719 1.581 .749

T3 1.141 .770 1.690 .511

Wealth quintile – Richest ®

Poorest .955 .556 1.640 .868

Poor .448 .253 .792 .006

Moderate .900 .571 1.419 .650

Rich .806 .525 1.237 .323

Constant .048

Note: ® — Reference Category

Table 11: Multivariate logistic regression on decision-making power on whom to marry among unmarried girls

Dependent variable – Girl having a major say on 95% C.I. whom to marry=1; rest=0 Sig.

Background characteristics Lower Upper Odds Ratio

Block – Nala ®

Mahagama .783 .488 1.258 .312

Godda .188 .101 .350 .000

Barhait/Barharwa .867 .532 1.413 .566

Jamtara .841 .530 1.333 .461

School status – Not in school ®

Currently in school 1.360 .960 1.926 .083

Age – 15-16 years ®

17-18 years 1.297 .954 1.765 .097

26 Dependent variable – Girl having a major say on 95% C.I. whom to marry=1; rest=0 Sig.

Background characteristics Lower Upper Odds Ratio

Mother's education – No schooling ®

At least 1 year of schooling 1.427 .995 2.046 .053

Father's education – No schooling ®

1-9 years of schooling 1.182 .801 1.745 .399

10 or more years of schooling 1.236 .758 2.017 .396

Mother's occupation – Unskilled worker ®

Skilled worker/ business/ salaried 1.308 .802 2.131 .282

Not working 1.155 .676 1.974 .599

Father's occupation - Not working or unskilled worker ®

Cultivator/Skilled worker 1.997 1.347 2.959 0.001

Business/Salaried 1.273 .798 2.028 .311

Religion – Non-Hindu ®

Hindu 1.978 1.341 2.920 .001

Caste – ST ®

SC 0.937 0.517 1.700 .831

OBC .762 .487 1.195 .237

Other .916 .506 1.657 .771

Type of village – T1 ®

T2 1.258 .836 1.893 .271

T3 1.390 .930 2.078 .109

Wealth quintile – Richest ®

Poorest .789 .450 1.382 .407

Poor .647 .378 1.107 .112

Moderate .833 .518 1.339 .450

Rich .951 .615 1.471 .822

Constant .052

27 Table 12: Attitudes toward child marriage among 15–18-year-old girls who are not yet married

Barhait/ Grand Mahagama Godda Jamtara Nala Barharwa total

Percentage who agree: Marrying girls young is likely to 79.2 81.0 78.1 81.2 79.4 79.8 have a negative impact on her education

Percentage who agree: A girl should be married only after she has been able to attain her educational and career 83.3 80.7 83.4 82.3 79.7 81.9 aspirations

Percentage who agree: It is appropriate for girls to express 44.4 48.7 44.1 38.7 40.4 43.1 displeasure if they do not like the partner chosen for them

Percentage who agree: Girls should be allowed to decide 69.6 60.2 78.7 71.5 67.0 69.5 when they want to marry

Percentage who agree: Girls should be allowed to say no 68.5 63.2 77.5 71.8 66.5 69.6 to an early marriage

Percentage who agree: Boys should be allowed to say no 67.9 64.4 78.4 70.4 62.9 68.8 to an early marriage

Percentage who agree: A girl should never be forced or 63.6 65.6 77.5 73.4 66.8 69.4 compelled into marriage

Percentage who disagree: Marrying girls young can help 54.0 72.1 55.3 57.5 70.1 61.6 protect the family honor/reputation

Percentage who disagree: Marrying girls young can help 71.5 84.9 72.5 77.4 82.4 77.6 resolve financial problems in the family

Percentage who disagree: Marrying girls young can 79.2 73.3 66.6 66.1 81.3 73.3 provide them security

Percentage who disagree: Marrying girls young can help prevent them from facing sexual violence, assault and 80.5 74.8 66.6 74.7 79.4 75.3 harassment

Percentage who disagree: Marrying girls young is preferable because younger brides are more obedient and 72.1 78.9 62.6 71.2 78.8 72.7 respectful of their husbands

Percentage who disagree: Younger brides require a lower 77.3 78.9 71.9 77.7 81.3 77.4 dowry than older brides

Percentage who disagree: Since girls have to get married, 66.0 78.0 73.3 76.6 76.1 74.0 they should not be sent for higher education

Percentage who disagree: After marriage, a woman should 40.8 53.7 47.5 51.6 46.7 48.0 work only if her in-laws and husband want her to

Percentage who disagree: Even if a girl does not want to be 35.6 51.9 39.3 41.1 47.5 43.0 married, she should honor the decisions/wishes of her family

Percentage who disagree: If the girl and boy choose each 61.4 65.3 57.3 62.1 61.8 61.5 other for marriage, they can get married irrespective of age

Total (N) 365 337 356 372 364 1794

28 Table 13: Attitudes of parents (father/mother/guardian) toward marriage

Barhait/ Grand Mahagama Godda Jamtara Nala Barharwa total

If a girl wants to marry a boy of her choice, what

should her family do?

Allow her to marry (%) 47.2 67.2 71.5 62.4 63.9 62.3

Not allow her to marry boy of her choice (%) 41.5 30.8 26.3 32.3 28.2 31.9

Throw her out of house (%) 0.5 1.0 0.5 0.2 0.0 0.4

Force boy to leave village (%) 1.2 0.0 0.2 0.0 2.0 0.7

Quickly marry her elsewhere (%) 9.2 1.2 1.5 4.6 5.9 4.5

Do you want your daughter to marry a boy of your/ your family’s choice or do you want her to have a love marriage?

Marry parents/family’s choice (%) 90.5 95.8 89.3 88.1 94.8 91.7

Love marriage (%) 8.8 3.5 7.6 11.9 4.0 7.2

The highest level of education a girl should complete 25.8 27.8 41.3 30.9 23.2 30.0 — graduation or above (%)

Right age for marriage

18.5 18.9 19.0 18.7 For girls – Mean (SD) 18.6 (1.5) 18.5 (1.8) (1.6) (2.2) (1.8) (1.8)

22.4 22.6 22.7 22.5 For boys – Mean (SD) 22.5 (2.6) 22.3 (2.7) (2.5) (3.1) (2.7) (2.7)

Total number of parents (N) 422 402 410 412 404 2050

29 Acknowledgements

This research report is a product of collective effort and research, with multiple rounds of feedback.

We thank all adolescent girls, boys and men, and parents who gave their time and participated in the survey.

We thank our ICRW Asia colleague, Prerna Kumar, for sharing initial inputs that helped us shape the report. We are indebted to members of the ICRW UMANG team — Rajendra Singh, Parasnath Verma, Nasreen Jamal, Tanvi Jha, Binit Jha, and Triloki Nath — for their contributions in the preparatory work, data collection, and analysis of the baseline survey.

We also acknowledge the efforts of the UMANG Project Coordinators and Block Coordinators from our partner organizations, SATHEE and Badlao Foundation. We thank data collection and quality assurance teams from Mind Field and SCHOOL, and quality assurance consultants, Mamta Srivastva and Shikha Srivastav for their efforts.

We thank Sandeepa Fanda for her support with program coordination and operations, and Arushi Sen and Ketaki V. Nagaraju for their editorial and design inputs.

30 ICRW Asia ICRW US C-59, South Extension, Part II, 1120, 20th St. NW, Suit 500 North, New Delhi - 110049 Washington, D.C. 20036 Tel.: + 91-11-46643333 Tel.: +202.797.0007 email: [email protected] email: [email protected] web.www.icrw.org/asia Web. www.icrw.org @ICRWAsia @ICRWDC @ICRWAsia @ICRW